Jean-Michaël Celerier authored 4 years ago and Albert Gräf committed 4 years ago

Else this does not seem to build with MSVC 2019

Addressed the issue where optimal zoom did not update the scrollable area and reenabled the zoom of the scrollbars when zooming (this scrollbar zoom size may require further reflection to decide the best path forward).

comment the line that defines an extra inlet for the 'inlet~' object

show error message in the case its a inlet(~) object but argument 'fwd' is not specified. Third approach

fix for case fwd message is not properly built, pd could crash

add zcheckgetfn method. given a selector and a list of param types, checks if the object has that method and if the params coincide

switch to mess3 method for calling fwd callback and add proper argument verification,  pack conditions to call inlet_fwd inside it, selector 'fwd' can be sent again through inlet objects

create wrappers of symbol and blob for inlet_fwd

create wrappers of pointer for inlet_fwd

We're not yet saving the hex colors as that will break in older patches.
But we do print out warning messages in this interim period and will
save the hex colors at a later date. Once that change is made, all the
interim versions will be able to read the new format

Vanilla commit 94d59de44ea23aa084596d1ba2050d3883511ec3

vanilla commit 853967de05253eaba8bf153450d573e8b6067576

Vanilla commit ab4750bd39964efd5c9a32fdaf9fb300d0d2f5d7

If a user added a device on the commandline, it's likely that they want
this device to be the default.

This allows Purr Data to work nicely inside a Flatpak sandbox, as we
can use `-alsaadd pulse` to enable audio output to PulseAudio by
default, with direct hardware access still an option.

This could lead to an assert failure inside ALSA if the output device
wasn't open and we tried to configure it anyway.

As the comment above the alsa_checkiosync() function suggests, there's
no sync to be done if we only have an input or an output.

The (input) and (output) markers were missing in some cases, and
incorrect in others.

- Disable touch-scrolling in situations where it interferes with other
  actions, like operating a slider, or dragging objects or the selection
  rectangle in edit mode.

- Truncate fractional touch coordinates which were causing glitches in the
  FUDI interface.

- Make dragging of iemgui labels and scalar draggables work (kind of).

TODO: Dragging (specifically in edit mode) can be a bit tricky in cases where
it's difficult to hit the right spot to initiate the drag action. This
*might* be solvable by making the algorithm identifying these spots a bit more
lenient, and accept touches which are a bit off.

Also, despite my best efforts, dragging iemgui label handles in edit mode
still needs quite some faffing about when using touch: After clicking the
iemgui object to make the handle appear, you first need to drag the handle a
little (horizontally, vertically, or diagonally) to make the corresponding
drag cursor appear, then release and touch the drag cursor *again* to drag the
handle to its final destination (and the label along with it). It goes without
saying that this seems overly complicated, but there doesn't seem to be an
easy way to work around this in the GUI, as the whole process goes back and
forth between GUI and engine in several steps in the current implementation.

Fixes the regression where the viewport is way too large after an optimal zoom operation (cf. #660).